phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James Taylor (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (PHOENIX-2724) Query with large number of guideposts is slower compared to no stats
Date Tue, 01 Mar 2016 22:57:18 GMT

    [ https://issues.apache.org/jira/browse/PHOENIX-2724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15174568#comment-15174568
] 

James Taylor commented on PHOENIX-2724:
---------------------------------------

It's good to understand our limits, but 876085 are too many guideposts and 1MB is too small
of a guidepost width. We should put some sanity checking around this. Same thing around the
check that determines if we run serially or parallelly wrt guidepost width. In particular,
a non aggregate query with no where clause shouldn't be chunked up in the same granularity
and shouldn't be run in parallel.

> Query with large number of guideposts is slower compared to no stats
> --------------------------------------------------------------------
>
>                 Key: PHOENIX-2724
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-2724
>             Project: Phoenix
>          Issue Type: Bug
>    Affects Versions: 4.7.0
>         Environment: Phoenix 4.7.0-RC4, HBase-0.98.17 on a 8 node cluster
>            Reporter: Mujtaba Chohan
>            Assignee: James Taylor
>
> With 1MB guidepost width for ~900GB/500M rows table. Queries with short scan range gets
significantly slower.
> Without stats:
> {code}
> select * from T limit 10; // query execution time <100 msec
> {code}
> With stats:
> {code}
> select * from T limit 10; // query execution time >20 seconds
> Explain plan: CLIENT 876085-CHUNK 476569382 ROWS 876060986727 BYTES SERIAL 1-WAY FULL
SCAN OVER T SERVER 10 ROW LIMIT CLIENT 10 ROW LIMIT
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message